Skip to Content
Oil refinery with flames from a flare stack at dusk

Motor Oil Group: Launching predictive maintenance in support of the digital refinery

Explore Motor Oil Group’s journey with SAP

Motor Oil Group wished to monitor its refinery equipment health and performance, reduce unexpected downtime, and lower maintenance costs. To gain a holistic view of its equipment, the company embraced machine learning and predictive analytics using SAP Business Technology Platform (SAP BTP). 



accuracy in explaining abnormal events from 120 to 20 hours in advance using root-cause analysis of historical data.


forward-looking time-series forecasting that enables accurate prediction of future sensor measurements.

Our initial predictive maintenance pilot is very accurate. With SAP Business Technology Platform, we have a complete view of our refinery equipment that will help us reduce unexpected downtime and lower maintenance costs.

Dimitrios Michalopoulos
Industrial Applications Head of IT Division
Motor Oil (Hellas) Corinth Refineries S.A. (Motor Oil Group)

The Challenge

Using machine learning and predictive analytics to reduce downtime

In any refinery, diligent equipment maintenance is critical to keeping workers safe and operations running at peak performance. Standard preventative maintenance processes only allow technicians to catch a potential problem when doing a scheduled maintenance review. 


Motor Oil Group wanted to do more than just prevent abnormal equipment behavior. It sought to harness sensor data to continuously monitor equipment health and predict potential malfunctions in advance. This would allow the company to avoid unexpected and costly shutdowns of key equipment and lower maintenance costs by addressing possible problems before they occur, so equipment and parts can be fixed rather than replaced. Ultimately, it wished to enhance its existing processes with data-driven predictive maintenance.

Creating a predictive model is an iterative process. By working with Accenture and the Data Science group at SAP, we are using SAP Business Technology Platform to dive deep into our equipment data and understand the right questions to ask to get the insight we need.

Dimitrios Michalopoulos
Industrial Applications Head of IT Division
Motor Oil (Hellas) Corinth Refineries S.A. (Motor Oil Group)

The Solution

Collaborating to create a predictive maintenance pilot based on SAP BTP

A data value workshop with the Data Science group at SAP and the refinery maintenance department helped identify opportunities to enhance existing maintenance processes through machine learning and predictive analytics. The use cases were prioritized based on the immediate business value they could provide. 


A successful proof-of-concept project analyzed four years of data on pressure, temperature, and vibration sensors from three compressors. The model was fed with the sensors’ alarm and trip thresholds to predict when these would be exceeded. 


Next, Motor Oil Group collaborated with SAP and Accenture’s Applied Intelligence unit in Greece to create a pilot architecture. The company used SAP HANA Cloud, application development services from SAP BTP and SAP Analytics Cloud to build predictive models for abnormal events based on sensor data and to provide the results through user-friendly dashboards and e-mail notifications. 

The Result

Gaining accuracy in understanding historical and future measurements

The predictive maintenance pilot on refinery-critical compressors yielded impressive results. Motor Oil Group achieved up to 77% accuracy for explaining abnormal events based on historical data and up to 70% accuracy for predicting future sensor measurements using root-case analysis and time-series analysis, respectively. The company intends on adjusting these models to increase effectiveness and then scaling to additional equipment assets. 


Overall, the company has gained a holistic view of its equipment and benefited from lower maintenance costs and less downtime due to the early notification of abnormal events. 


Plus, with a system that learns from its own data, the business can expect the predictions to become more accurate and effective over time. 

SAP helps Motor Oil Group run better

Key business outcomes and benefits

  • Real-time equipment health and performance monitoring and the ability to predict future behavior
  • Sensor-level time-series forecasting that helps predict a sensor’s behavior over the coming 24 hours
  • Strong predictability for abnormal events across groups of equipment
  • Holistic view of equipment that provides a better understanding of event triggers and overall decomposition
  • Deep-dive equipment analysis that reveals hidden patterns and provides insight about how various elements and events affect each other

Featured solutions and services

  • SAP Business Technology Platform brings data management, analytics, AI, application development, automation, and integration into one, unified environment.
  • SAP HANA Cloud provides advanced analytics on multimodel data, on premise and in the cloud.
  • SAP Analytics Cloud combines business intelligence, augmented and predictive analytics, and planning capabilities into one cloud environment, supporting advanced analytics.

About Motor Oil Group

Founded in 1970, Motor Oil (Hellas) Corinth Refineries S.A. (Motor Oil Group) is one of the industry leaders in crude oil refining and the sales of petroleum products across Greece and the Eastern Mediterranean region. Its refinery business is complex, and includes widespread ancillary plants and fuel distribution facilities that export to more than 45 countries.

Featured partner

Accenture plc is a multinational professional services company that specializes in IT services and consulting. It has just shy of 700,000 employees worldwide and 7,000 clients served throughout more than 120 countries. It works with 185 partners as well as 89 of the Fortune Global 100. 

Back to top